Deregulation of glucose and lipid metabolism in peripheral tissues is a hallmark of type 2 diabetes. AMP- activated protein kinase (AMPK) is a master regulator of cellular and organismal metabolism controlling glucose and lipid homeostasis. AMPK is activated by low nutrients, exercise, adipokines, and by the widely used diabetes therapeutics metformin and the thiazolidinediones (TZDs). In response to these stimuli, AMPK acts in the liver to reduce gluconeogenesis and lipogenesis through poorly understood mechanisms. We, and others, identified the serine/threonine kinase LKB1 as the critical upstream kinase mediating AMPK activation. We subsequently created a genetic deletion of LKB1 in the liver of adult mice, which resulted in complete loss of hepatic AMPK activity and dramatically increased gluconeogenesis and hepatic lipid accumulation. Using these mice, we demonstrated that LKB1 was required in liver for metformin to lower blood glucose levels, the first genetic proof of a specific pathway being required for the therapeutic action of metformin. The next big question is to understand how metformin impinges on LKB1/AMPK signaling and how they in turn regulate glucose metabolism. We propose to determine the role of LKB1 and AMPK in the control of hepatic glucose metabolism and in the therapeutic action of metformin and TZDs. First, as LKB1 is known to activate 12 AMPK-related kinases in addition to AMPK, we will determine whether loss of AMPK alone mimics the effects of loss of LKB1 on glucose metabolism and the response to metformin or TZDs. To this end, we will conditionally delete LKB1 or both catalytic AMPK1 genes in the liver of adult mice. Second, using hepatocytes derived from these mice and RNAi, we will define the critical upstream and downstream components of the LKB1/AMPK pathway required for the regulation of specific metabolic processes. Finally, we will identify a number of new effectors of AMPK that control metabolism using a combination of unique proteomic approaches to purify novel AMPK substrates in addition to transcriptional profiling in our genetically defined cells following metformin or TZD treatment. These studies will better illuminate the mechanism of action of these two widely used type 2 diabetes modalities, as well as identifying many new targets for the development of future therapeutics.